1
Obstructed View:
New York State Aorney General
Eric T. Schneiderman
From the Office of:
What’s Blocking New Yorkers from Geng Tickets
SolD oUt AgAIn|
CAPTCHA
Type the two words:
1
This report was a collaborave eort prepared by the Bureau of Internet and Technology
and the Research Department, with special thanks to Assistant Aorneys General Jordan
Adler, Noah Stein, Aaron Chase, and Lydia Reynolds; Director of Special Projects Vanessa Ip;
Researcher John Ferrara; Director of Research and Analycs Lacey Keller; Bureau of Internet and
Technology Chief Kathleen McGee; Chief Economist Guy Ben-Ishai; Senior Enforcement Counsel
and Special Advisor Tim Wu; and Execuve Deputy Aorney General Karla G. Sanchez.
2
TABLE OF CONTENTS
Execuve Summary ....................................................................... 3
The History of, and Policy Behind, New York’s Tickeng Laws ....... 7
Current Law ................................................................................... 9
Who’s Who in the Tickeng Industry ........................................... 10
Findings ....................................................................................... 11
A. The General Public Loses Out on Tickets to
Insiders and Brokers .................................................................... 11
1. The Majority of Tickets for Popular Concerts Are Not Reserved
For the General Public .......................................................................... 11
2. Brokers & Bots Buy Tickets in Bulk, Further Crowding Out Fans ...... 15
a. Ticket Bots Amass Hundreds or Thousands of Tickets in an Instant ....... 15
b. Even without Bots, Brokers Use Industry Knowledge and Relaonships
to Gain an Edge Over Fans ...................................................................... 21
c. Ticket Limits Are Not Regularly Enforced ................................................ 22
d. Brokers, With and Without Bots, Buy Up Many
Tickets to New York Events ..................................................................... 23
e. Brokers Sell Tickets at Substanal Markups ............................................ 25
f. Speculave Tickets Increase Confusion and Risk for Fans ........................ 26
g. The Market Structure Is Atypical ............................................................. 27
B. High Fees for Unclear Purposes Raise Concerns ......................... 28
1. Fees are Charged by Ticket Vendors and Venues ............................. 29
2. Event Ticket Fees are Higher than Most Online Vendors ................. 31
C. Restraints of Trade in Tickeng ................................................... 32
1. Seng of Ticket Resale Price Floors ................................................. 32
2. Pracces That Impede Consumer Access to
Alternave or “Unocial”Ticket Resale Plaorms................................ 33
Recommendaons ....................................................................... 34
Appendix – Methodology for NYAG Analyses ...............................38
3
EXECUTIVE SUMMARY
The New York Aorney General (“NYAG”) regularly receives complaints from New Yorkers
frustrated by their inability to purchase ckets to concerts and other events that appear to sell
out within moments of the ckets’ release. These consumers wonder how the same ckets
can then appear moments later on StubHub or another cket resale site, available for resale at
substanal markups. In response to these complaints, NYAG has been invesgang the enre
industry and the process by which event ckets are distributed from the moment a venue
is booked through the sale of ckets to the public. This Report outlines the ndings of our
invesgaon.
More than 15 years ago, NYAG issued a landmark report on what it called “New York’s largely
underground and unexamined cket distribuon system. The report announced that it was
a system that provides “access to quality seang on the basis of bribes and corrupon at the
expense of fans.
Since that report was wrien broad changes in the technology of ckeng and an overhaul of
New York’s ckeng laws have completely transformed the landscape, in ways both good and
bad. Following the repeal of New York’s an-scalping laws” in 2007, the once underground
cket resale economy moved parally above ground. This change has produced some benets:
online marketplaces have replaced waing in long lines, the growth of cket resale plaorms
has somemes made it easier to sell unwanted ckets, and the last minute-minded can aend
shows without interacng with potenally dishonest street scalpers.
Yet many of the problems described in 1999
have persisted and, in some cases, have grown
worse. Whereas in many areas of the economy
the arrival of the Internet and online sales has
yielded lower prices and greater transparency,
event ckeng is the great excepon. The
complaints NYAG receives from consumers
concerning ckeng commonly cite “price
gouging, scalping, outrageous fees” and
“immediate sell-outs.” As one cizen wrote,
in a typical complaint: The average fan has
no chance to buy ckets at face value this
is a disgrace. Many performers voice similar
frustraons.
1
The problem is not simply that demand for prime
seats exceeds supply, especially for the most in-
demand events. Tickeng, to put it bluntly, is a
xed game. Consider, for example, that on December 8, 2014, when ckets rst
went on sale for a tour by the rock band U2, a single broker purchased 1,012 ckets to one show
at Madison Square Garden in a single minute, despite the cket vendors claim of a “4 cket
1. See, e.g., Mark Savage, “Sir Elton John: Secondary ticket prices disgraceful’,”BBC News (Dec. 16, 2015),
available at
http://www.
bbc.com/news/entertainment-arts-35091230; Patrick Doyle, “Eric Church on Scalpers, Bro-Country and Blake Shelton Scandal,
Rolling Stone (June 11, 2014),
available at
http://www.rollingstone.com/music/news/eric-church-on-scalpers-bro-country-and-
blake-shelton-scandal-20140611.
“I think its extoronate
and I think its disgraceful.
- Sir Elton John, referring to inated
secondary sales
“Its a systemac problem
with music ...
- Eric Church, referring to scalpers
4
limit.By the end of that day, the same broker and one other had together amassed more than
15,000 ckets to U2’s shows across North America.
Consider that brokers somemes resell ckets at margins that are over 1,000% of face value.
Consider further that added fees on ckets regularly reach over 21% of the face price of ckets
and, in some extreme cases, are actually more than the price of the cket. Even those who
intend their events to be free, like Pope Francis, nd their good intent defeated by those who
resell ckets for hundreds or even thousands of dollars.
Findings
A. The General Public Loses Out on Tickets to Insiders and Brokers.
New Yorkers keep asking the same queson: why is it so hard to buy a cket at face value?
1. Holds & Pre-Sales Reduce the Number of Tickets Reserved for the General Public.
Our invesgaon found that the majority of ckets for the most popular concerts are not reserved
for the general public at least in the rst instance. Rather, before a member of the public can buy
a single cket for a major entertainment event, over half of the available ckets are either put on
“hold” and reserved for a variety of industry insiders including the venues, arsts or promoters,
or are reserved for “pre-sale” events and made available to non-public groups, such as those
who carry parcular credit cards.
2. Brokers Use Insider Knowledge and Oen Illegal Ticket Bots to Edge Out Fans.
When ckets are released, brokers buy up as many desirable ckets as possible and resell them
at a markup, oen earning individual brokerages millions of dollars per year. To ensure they get
the ckets in volume, many brokers illegally rely on special soware – known as Ticket Bots – to
purchase ckets at high speeds. As the New York Times reported, Ticketmaster has esmated
that “60 percent of the most desirable ckets for some shows” that are put up for sale are
purchased by Bots.
2
Our research conrms that at least tens of thousands of ckets per year are
being acquired using this illegal soware.
Brokers then mark up the price of those ckets by an esmated 49% on average, but somemes
by more than 1,000% – yielding easy prots. In at least one circumstance, a cket was resold at
7,000% of face value. Finally, some brokers sell speculave ckets,meaning they sell ckets
that they do not have but expect to be able to purchase aer locking in a buyer. Speculave
ckets are a risk for consumers and also drive up prices even before ckets are released.
NYAG’s invesgaon idened those brokers re-selling the most ckets for New York events.
Nearly all were unlicensed, and several employed illegal Ticket Bots to buy ckets. A number of
specic invesgaons and enforcement acons are in process.
2. Ben Sisario, “Concert Industry Struggles With ‘Bots That Siphon O Tickets, N.Y. Times (May 26, 2013),
available at
http://www.nytimes.com/2013/05/27/business/media/bots-that-siphon-o-tickets-frustrate-concert-promoters.html
5
B. High Fees for Unclear Purposes Raise Concerns.
Another common complaint concerns the unclear and unreasonable service fees” added to the
face value of ckets, which are generally set by venue operators and cket vendors. New York
law prohibits these pares from adding fees to the prices of ckets unless they are connected
to the provision of a “special service” and are “reasonable. Our examinaon of cket fees
set by 150 venues in New York raises concerns, revealing that unclear convenience charges,
service fees,” and “processing fees” somemes reach outlandish levels, either as a percentage
of the ckets face value or in absolute dollar terms. On average, New York venues and their
ckeng vendors charge fees averaging 21% of face values, which exceeds what other online
sellers charge. Moreover, we found fees as high as $42 aached to a cket to see Professional
Bull Riding at Madison Square Garden and $28 to see Janet Jackson at Jones Beach Theater.
C. Restraints of Trade Exist.
NYAG is concerned by the growing imposion of resale price oors (i.e. “no sales below list
price”), along with eorts to mandate that ckets be sold on a single “walled garden” market, as
opposed to consumers having the opon of buying ckets from dierent resale plaorms. We
are also interested in the degree to which excessive service charges may constute evidence of
abuse of monopoly power, especially as they relate to the resale of sports ckets.
Recommendaons
In 1999, NYAG found that, despite long-standing regulaon, the law “has not succeeded in
eliminang the abuse it was intended to address. That remains true today, and New York
remains in need of greater protecons for the buying public. We therefore oer several
recommendaons:
A. Ticket Resale Plaorms Must Ensure Brokers Comply With the Law.
Ticket resale plaorms are in the best posion to ensure that their broker customers follow the
law, and they must take meaningful steps to do so. Specically, these plaorms should require
that brokers provide their New York license numbers as a condion of using the resale plaorm,
and disclose to potenal customers the face value of ckets they are oering for sale, as already
required by New York law.
B. Industry Players Must Increase Transparency Regarding
Ticket Allocaons and Limits.
The industry must provide greater transparency into the allocaon of ckets, to increase
accountability and enable the public to make informed choices. Promoters of events, who
know the number of seats being held, should provide that informaon to cket vendors, such
as Ticketmaster, to make available to the public. In addion, wherever cket vendors claim that
cket limits are enforced, they should enforce those limits as a maer of course on a per-person
basis. If such limits are not actually being enforced, cket vendors must make that clear.
6
C. Ticket Vendors Must Address the Bot Epidemic.
Bot use is a major reason why New Yorkers cannot get ckets at face value. While the industry
works on long-term technological soluons to this problem, steps can be taken to reduce Bot
use in the near-term. NYAG has contacted Ticketmaster and another major cket vendor, AXS,
to discuss concrete reforms, such as preempve enforcement of cket limits, analyzing purchase
data to idenfy ongoing Bot operaons for prosecuon, and invesgang resellers of large
volumes of ckets to popular shows, among others.
D. The Legislature Should Act.
While there is no reason the industry should wait for legislave acon to implement the reforms
outlined above, the Legislature should act to ensure that reform is meaningful and lasng.
Specically:
i. Mandate the industry reforms outlined above.
ii. End the ban on non-transferrable paperless ckets.
A soluon that most industry parcipants agree is eecve at reducing broker acvity is
the use of non-transferrable “paperless ckets.” Unlike paper ckets and electronic ckets
that are freely transferrable from the buyer to another person, non-transferrable paperless
ckets require an event aendee to present the credit card that was used to purchase the
cket. As a result, the inial purchaser typically must be present to use the cket. State
law creates a
de facto
ban on these paperless ckets, but this rule makes New York an
outlier – ours is the only state that bans the pracce – and this ban should be repealed.
iii. Impose criminal penales for Bot use.
Given that cket resellers are making considered business decisions when they deploy
Bots to acquire massive amounts of ckets near-instantaneously, the prospect of criminal
prosecuon may well have a deterrent eect on this conduct.
iv. Cap permissible resale markups.
Unl 2007, New York capped the markup resellers could charge, and the State removed
that cap in hopes of beneng consumers. Unfortunately, compeon-driven savings
intended to benet fans have instead been converted to prots for a handful of savvy
middlemen using mulple employees or computers, or illegal Bots. New York should
reinstute a reasonable limit on resale markups.
7
THE HISTORY OF, AND POLICY BEHIND,
NEW YORK’S TICKETING LAWS
New York’s approach to ckeng has undergone dramac changes over the last een years.
Understanding both the history and the policy goals of previous and current ckeng legislaon
is important to all that follows.
3
1920 – 2007: The “An-Scalping” Era.
The greatest evil that theatergoers in this city have to contend with is the cket speculator,
wrote a New York City magistrate in 1901. They are praccally highwaymen and hold up
everybody that goes to a place of amusement.
4
As the quote suggests, the resale of ckets for
prot has long been considered an aggravaon in this State. In 1922, Governor Nathan Miller
declared the need for acon to combat what he called gross proteeringby “cket scalpers.
He signed a law, widely described as an “an-scalping law,that began the State’s regulaon of
cket sales for major events. It originally regulated “boxing and wrestling, relang to ckets to
places of entertainment,” and capped the price of resale ckets at two dollars above face value.
5
The law maintained its basic character for seventy years, broadening in scope, and adding various
provisions to ban bribes paid to venues by brokers for ckets (known as “ice”). Eventually, the
two-dollar limit became a 20% or 45% price cap (depending on the size of the venue), which
the law required to be printed legibly on each cket.
6
Over the years, the legislature periodically
imposed stricter penales on those selling ckets near venues and took steps to give the State
the power to police out-of-state actors dealing in ckets to events within New York.
Nonetheless, as NYAG’s report made clear in 1999, the law was both inconsistently enforced and
dicult to enforce. Underground brokers ourished, oen openly oung the law. By that year,
cket brokering had grown into a large underground economy typied by law-breaking, secret
cket holds, bribe paying, and a general state of corrupon at all levels.
2007 – Present: A Regulated Industry.
In 2007, in a sweeping legal change, the New York legislature decriminalized all resale of ckets
for prot, or “scalping,and adopted an approach that sought to treat ckeng as a regulated
industry. With the price caps lied, brokers were now free to operate openly and sell ckets at
whatever prices consumers might pay. In the face of crics who warned that prices would rise
and brokers would reap unseemly prots, the sponsors of the Bill expressed some hope that the
new approach might, in fact, reduce cket prices on the resale markets. As Assemblyman Joseph
D. Morelle argued at the me, “By allowing greater compeon for the resale dollar, we may
actually see a decrease in secondary prices, which is … ulmately best for the consumer.
7
3. The governing statutory provisions are found in New York Arts and Cultural Aairs Law, Article 25 (“N.Y. Arts & Cult. A. L.”).
4. Kerry Segrave, Ticket Scalping: An American History, 1850-2005 (2006), p. 55.
5. N.Y. General Business Law §§ 167-69k (McKinney 1922) (repealed 1983).
6. The following notice was required to be printed on tickets:
“If the venue to which this ticket grants admission seats 6000 or fewer persons, this ticket may not be resold for more than 20%
above the price printed on the face of this ticket, whereas if the venue to which this ticket grants admission seats more than 6000
person, this ticket may not be resold for more than 45% above the price printed on the face of this ticket.
7. Oce of Joseph D. Morelle, “Morelle Bill Establishing Free Market for Ticket Resellers Passes Assembly” (May 29, 2007),
available at
http://assembly.state.ny.us/mem/Joseph-D-Morelle/story/22938.
8
The operang premise of the 2007 law was not to completely deregulate cket resale, but
to legalize it pursuant to regulaon and taxaon. Pre-2007, many of the brokers operated in
open violaon of the law; the 2007 law introduced a revised licensing system supervised by
the Secretary of State. The licensing system required, among other things, various disclosures
of ckets sold, the posng of a bond to cover counterfeit ckets, and the payment of a $5,000
annual fee.
Above all, the 2007 law represented
a new and experimental approach
to the cket industry, made clear by
the fact that the law regularly sunsets
and is therefore subject to ongoing
nkering by the legislature. The rst
adjustments came in 2010, when the
law underwent major changes, three
of which are important here. First,
for the rst me, the law banned
the use of cket-buying soware,
or “Bots. The New York Senate
commentated that such soware is
“used by unscrupulous speculators to
purchase ckets at inial sale ahead
of consumers intending to aend an
event, and put enforcement dues
in the hands of the Aorney General.
8
Evident in the ban on Bots and the
legislature’s commentary is the idea that cket buying, when presented to the public as a fair
contest, should in fact be fair, and not subject to manipulaon by soware or similar techniques.
In other words, the legislature prefers that the ability to obtain ckets be based on factors like
showing up at the right me or simple luck of the draw, as opposed to gaining an advantage
through soware that funcons dierently than a human could.
Second, an important 2010 legal reform addressed growing service fees imposed by primary
cket sellers and their cket vendor agents, such as Ticketmaster or Telecharge. It mandated, for
the rst me, that any fees charged for “special services” must be “reasonable.
9
Third, the law
barred non-transferrable paperless ckets, as discussed in greater detail later in this Report.
10
The clear tenor of the ckeng statute is to make cket brokering into a regulated industry and
connue safeguarding the public against fraud, extoron, and similar abuses. Ticket brokers
are not illegal, but must obtain a license, post bonds, disclose their acvies to the Secretary of
State, and avoid using prohibited Bots. Whether the new system has operated in the interests
of New York cizens remains an open queson of public policy.
8. New York State Senate Introducer’s Mem., SB No. 3840-A.
9. N.Y. Arts & Cult. A. L. § 25.29 (as amended July 2, 2010).
10. Id. § 25.30.
Whats a Bot?
A Ticket Bot is soware that automates
cket-buying on plaorms such as
cketmaster.com.
Automaon lets the Bot (1) perform
each transacon at lightning speed, and
(2) perform hundreds or thousands of
transacons simultaneously.
As a result, in the rst moments aer
ckets to a top show go on sale, Bots
crowd out human purchasers and can
snap up most of the good seats.
9
CURRENT LAW
New York State law prohibits engaging “in the business of reselling any ckets to a place of
entertainmentwithin New York State without rst procuring “a license.
11
Those who resell, oer
to resell, or purchase with the intent to resell ve or more ckets without a license are guilty
of a misdemeanor and subject to penales.
12
Addionally, the law requires that brokers post
a $25,000 bond with their license applicaon to ensure compliance with the law’s provisions
and cover damages to their customers from any misconduct.
13
Brokers that resell ckets online
must also display a hyperlink to a copy of their licenses,
14
and they must display the face value of
ckets along with the resale prices.
15
New York State law also makes it illegal for brokers to use Bots to bypass security measures on
the websites of cket vendors, such as Ticketmaster, or to maintain any interest in or control of
such Bots.
16
Anyone who violates these provisions is subject to penales and the forfeiture of
prots.
17
No current federal law specically prohibits the use of Ticket Bots, but legislaon has been
proposed that would do so. The Beer On-line Ticket Sales Act of 2014 or the BOTS Act, supported
by Senator Charles Schumer, would prohibit, as an unfair and decepve act, the sale or use of
Ticket Bots. It would also provide for criminal penales for those who engage in this conduct.
11. Id. § 25.13.
12. Id. §§ 25.09(2), 25.35.
13. Id. § 25.15.
14. Id. § 25.19.
15. Id. § 25.23.
16. Id. § 25.24.
17. Id.
10
WHO’S WHO IN THE TICKET INDUSTRY







 




















11
FINDINGS
A. The General Public Loses Out on Tickets to Insiders and
Brokers.
The majority of ckets for popular concerts are diverted away from the general public. First,
various “hold” and “pre-sale” programs reduce the total number of ckets available to the
general public. Then, of those ckets that remain, brokers use a number of methods including
Bots to seize a large poron of the remaining ckets. Brokers even sell ckets that they do not
actually own, known as “speculave” ckets, to fund their later purchases.
1. The Majority of Tickets for Popular Concerts Are Not Reserved
For the General Public.
It may come as no surprise to fans that many concert ckets are never made available to the
general public, but there is lile informaon available as to where precisely those ckets actually
go. To bring some light to this issue, NYAG has obtained and analyzed informaon regarding the
allocaon and distribuon of ckets from the largest promoters, Live Naon and AEG, for the
top-grossing shows in New York for the years 2012-2015.
NYAG found that, on average, only about 46% of ckets are reserved for the public. The remaining
54% of ckets are divided among two groups: holds (16%) and pre-sales (38%).
18
Holds are ckets
that are reserved for industry insiders, such as arsts, agents, venues, promoters, markeng
departments, record labels, and sponsors. Pre-sales make ckets available to non-public groups
before they go on sale to the general public. The two most common pre-sale events are for
credit card holders and members of fan clubs, but venues, promoters, or other groups may run
pre-sale events as well. Brokers, as discussed below, oen begin harvesng ckets at these pre-
sale events and later during the general on-sale.
18. The methodology for this analysis is in the Appendix.
12
Figure 1. Less Than Half of Tickets Are Reserved for the General Public.
Source: Live Naon/Ticketmaster (2012 – 2013) and AEG (2012 – 2015)
19
Among the top grossing shows in New York, approximately 16% of all ckets were reserved
through “holds” for industry insiders: arsts, venues, agents, markeng departments, sponsors,
promoters, and execuves.
20
Oen, venues are the biggest beneciaries of these holds. For
example, Madison Square Garden was typically allocated more than 900 ckets per concert
and Barclays Center was allocated an average of 500 ckets per concert for the highest grossing
concerts held at those venues.
For some concerts, cket holds account for well over 16% of the available ckets. For instance, a
Kanye West show at Barclays Center held 29% of ckets for insiders, including over 2,000 ckets
for the promoter and more than 500 ckets for the venue. In all, fourteen of the State’s most
popular shows held at least 20% of available ckets for insiders, as shown in Figure 2. Holding
a large number of ckets is not necessarily a pracce among all shows – the Vans Warped Tour
2013 and Barry Manilow concerts, both at the Nassau Coliseum, held 5% or less of ckets.
19. The methodology for this analysis is in the Appendix.
20. It is worth noting that a hold simply represents a reservation, and that in most cases a group that has been allocated tickets
through a hold will not use a portion of those tickets. In these instances, the tickets are released back to the promoter, who will
typically make the tickets available for sale to the public. Because the tickets are usually not released through a publicized on-sale
event, tickets released in this manner are often purchased by brokers who are constantly searching for new tickets that are made
available closer to the date of the event.
Ticket Allocaon
Holds for Insiders
Pre-Sale Events
General Public
13
Figure 2. Fourteen Popular Shows Held at Least 20% of Tickets for Insiders.
Source: Live Naon/Ticketmaster (2012 – 2013) and AEG (2012 – 2015)
21
Filtered for shows with 20% or more total capacity placed on hold
Approximately 38% of ckets to the most popular shows in New York were reserved for “pre-
sales, advance sales to select groups of fans and cardholders of major banks and nancial
instuons.
22
These nancial instuons, including American Express, Cibank, and Chase, have
negoated agreements whereby their cardholders have the opportunity to purchase ckets
21. The methodology for this analysis is in the Appendix.
22. For the top shows, it is often the case that all of the tickets that are made available through the pre-sale events are sold. For
less popular shows, tickets set aside for pre-sale events can go unsold. When tickets set aside for a pre-sale event are not sold, the
unsold tickets are reallocated to the initial public on-sale.
Trey Songz
Madison Square Garden
(Theater) on 3/1/2012
20%
Jusn Bieber
Madison Square Garden on
11/29/2012
28%
Jusn Bieber
Madison Square Garden on
11/28/2012
28%
Keith Urban
Madison Square Garden on
1/29/2014
20%
The Voice
Beacon Th eatre on 7/9/2014
22%
Ricardo Arjona
Madison Square Garden on
2/26/2012
23%
Fresh Beat Band
Madison Square Garden
(Theater) on 11/30/2013
24%
Fresh Beat Band
Madison Square Garden
(Theater) on 11/29/2013
24%
Kanye West
Barclays Center on
11/20/2013
21%
Kanye West
Barclays Center on 11/19/2013
29%
Drake
Jones Beach on 6/16/2012
20%
Luke Bryan
Nassau Coliseum on
2/7/2013
20%
Farm Aid
Saratoga Perf orming Arts
Center (SPAC) on 9/21/2013
22%
Fleetwood Mac
Madison Square Garden on
4/8/2013
24%
NKOTB
Nassau Coliseum on 6/1/2013
24%
Promoter
LiveNao n
AEG
14
before the ckets are released for sale to the general public. Importantly, fans are not necessarily
the ones purchasing ckets released in pre-sale events. For popular events, brokers, realizing the
advantages of pre-sale programs, purchase heavily and oen use illegal Bots to maximize their
yield in pre-sales, as discussed in greater detail below.
23
While the majority of pre-sale events
are conducted by credit card companies, members of other groups, including fan clubs, social
media websites, and specialty shopping sites also receive advance opportunies to purchase
ckets.
The number of ckets reserved for pre-sale events can be quite large for some concerts. For
example, for more than 30 of New York’s most popular concerts, at least half of the ckets were
set aside for pre-sale events. Figure 3 idenes ten shows that made more than half of the
ckets available through pre-sale events where none of the ckets were explicitly reserved for
fan clubs.
Figure 3. Ten Popular Shows Reserved Over 50% of Tickets
Through Pre-Sales Events, None Explicitly Earmarked for Fan Clubs.
Source: Live Naon/Ticketmaster (2012 – 2013)
24
Filtered for shows with more than 50% of total capacity sold through
pre-sale events and no ckets specically earmarked for fan clubs
23. Even setting aside the problem of brokers, cardholder pre-sales may disadvantage many fans. They reserve tickets for hold-
ers of select credit cards, who may generally be wealthier than fans who lack those cards and the benets that come with being
a cardholder. As a result, these pre-sales give wealthier fans a better chance of getting tickets at face value, while decreasing the
supply of face-value tickets available to less wealthy fans.
24. The methodology for this analysis is in the Appendix.
Fleetwood Mac
Madison Square Garden on 4/8/2013
61%
Coldplay & Jay-Z
Barclays Center on 12/31/2012
70%
Coldplay &
Jay-Z
Barclays Center on 12/30/2012
52%
Jay-Z & Jusn
Timberlake
Yankee Stadium on 7/20/2013
69%
Jay-Z & Jusn
Timberlake
Yankee Stadium on 7/19/2013
71%
SteelyDan
Beacon Theatre on 10/8/2013
64%
SteelyDan
Beacon Theatre on 10/5/2013
65%
SteelyDan
Beacon Theatre on 10/4/2013
64%
SteelyDan
Beacon Theatre on 10/1/2013
64%
SteelyDan
Beacon Theatre on 9/30/2013
64%
Promoter
LiveNao n
15
Holds and pre-sales can leave few ckets for the public. Indeed, for many of the top shows, less
than 25% of ckets were actually released to the general public in an inial public on-sale. For
example, just over 1,600 ckets (12% of all ckets) were released to the public during the inial
public on-sale for a July 24, 2014 Katy Perry concert at Barclays Center. Similarly, for two Jusn
Bieber concerts at Madison Square Garden, on November 28, 2012 and November 29, 2012,
fewer than 2,000 ckets (15% of all ckets) to each show were released to the public during the
inial public on-sale.
2. Brokers & Bots Buy Tickets in Bulk, Further Crowding Out Fans.
The broker industry, while not the explicitly underground industry it was in the 20th century,
remains dicult to learn about and oen operates in violaon of one or more State laws.
The average fan vying to purchase a cket to a popular concert has lile hope of compeng
against brokers, many of whom use illegal and unfair means to purchase ckets. Brokers are
able to purchase large quanes of ckets to popular events by (a) employing illegal Bots, and/
or (b) exploing their industry knowledge and relaonships to get access to ckets.
a. Ticket Bots Amass
Hundreds or Thousands
of Tickets in an Instant.
A Ticket Bot is a soware
program that automates
the process of searching
for and buying ckets to
events on cket vendor
plaorms, such as
Ticketmaster. Using Bots,
a broker can automate the
process of searching for
and buying ckets so that it occurs in the blink of an eye, and conduct tens, hundreds, or even
thousands of lightning-fast transacons at the same me.
To successfully buy thousands of ckets, Bots must perform four dierent funcons. These
funcons may be handled by one Bot or several.
First, spinner, or “drop checker,” Bots constantly monitor ckeng sites to detect the release, or
drop,of ckets. Ticketmaster has stated that spinners can account for as much as 90% of the
trac to its website.
25
Second, Bots automate the search for and reservaon of ckets that are up for sale. Here,
brokers exploit the fact that most ckeng plaorms “reserve” ckets for a few minutes to give
the (human) customer me to complete the purchase. Brokers take advantage of this feature
by conducng mulple near-instantaneous searches at the precise moment ckets are released
for sale, throwing hundreds of ckets into reserve and removing them from the pool of ckets
25. Ticketmaster L.L.C. v. RMG Technologies, Inc., 07-Civ.-2534 (C.D. Cal.) (Application for Entry of Default Judgment); Ben Sisario &
Emily B. Hager et al., “Fair Ticketing: Fans Before Scalpers, N.Y. Times Video (May 27, 2013),
available at
http://nyti.ms/13V6sdO.
“...According to Ticketmaster, bots have been used to
buy more than 60 percent of the most desirable tickets
for some shows….”
Concert Industry Struggles With ‘Bots’ That
Siphon Off Tickets
16
available for purchase. The broker may then review the search results and choose which seats
to buy.
26
Figure 4 shows an example of a Bot that has searched for and reserved several groups
of ckets, displaying the groups and the me remaining for the Bot-user to purchase each one.
Figure 4. Bot Displays Time Remaining to Buy
Temporarily “Reserved” Tickets.
Source: Ticketbots.net
Third, Bots automate the process of purchasing ckets, using dozens or hundreds of purchaser
names, addresses, and credit card numbers. The names of the “buyers” may be simply invented,
or they may be borrowed from real people.
Fourth, and nally, Bots must defeat the an-Bot security measures that most cket-selling
plaorms employ. The most obvious of the cket vendors’ defensive tools that Bots must bypass
is the “CAPTCHA” test (“Completely Automated Public Turing test to tell Computers and Humans
Apart”). This test usually requires that a user prove she is human by reviewing text that is
distorted or obscured and entering those characters into a text box.
26. Some brokers use this interval to oer the temporarily reserved tickets for sale on secondary market platforms such as Stub-
Hub at a given markup, and only if the resale is quickly consummated do they then actually buy the reserved tickets from the
primary ticket vendor. This ability to use temporary reserve to avoid risk greatly undermines a shibboleth repeated to NYAG
during our investigation, that brokers benet the ticket industry as a whole, including artists, promoters, and venues, by taking on
nancial risk through up-front purchases of lots of tickets which they may be unable to resell.
17
Figure 5. Ticket Vendors Use Dierent Types of CAPTCHA.
Unfortunately, many Bot programmers have been able to bypass these forms of CAPTCHA.
Over the years, Ticketmaster has repeatedly refurbished its CAPTCHA program, using dierent
versions of CAPTCHA created by third pares such as Google and Solve Media.
27
While these
CAPTCHAs may have stopped some Bots, more sophiscated Bot programmers quickly adapted.
In many cases, they collected thousands of the new CAPTCHAs and used them to “train” their
soware to “read” the new CAPTCHAs through improved opcal character recognion. In other
instances, the Bots transmit in real-me images of the CAPTCHAs they encounter on Ticketmaster
and other sites to armies of “typers,human workers in foreign countries where labor is less
expensive. These typers employed by companies such as Death by CAPTCHA, Image Typerz,
and DeCaptcher – read the CAPTCHAs in real-me and type the security phrases into a text box
for the Bot to use to bypass cket vendors’ defenses and use their sites.
28
Addionally, at various points, Ticketmaster abandoned CAPTCHAs for its mobile app on the
iPhone plaorm, based on the premise that it could beer verify idenes on mobile and
therefore did not need a CAPTCHA test. Our invesgaon of brokers reveals that some have
targeted the CAPTCHA-free mobile plaorm by either designing Bots to mimic mobile devices
or by using an arsenal of actual mobile devices to buy ckets.
27. Some CAPTCHAs incorporate technology that attempts to identify Bots before an image is even shown, to present an easier
test to likely humans and more dicult CAPTCHAs to suspected Bots. See, e.g., Solve Media – High Security Standards, http://
solvemedia.com/security/index.html. Many sophisticated Bot developers have bypassed these additional measures.
28. Some CAPTCHAs now do away with security phrases and use other types of puzzles. See, e.g., Google – “Introducing No CAPT-
CHA reCAPTCHA, https://googleonlinesecurity.blogspot.com/2014/12/are-you-robot-introducing-no-captcha.html. It remains to
be seen if these new approaches will be harder to bypass, especially for Bots that use teams of human typers.
18
To eecvely beat out the many fans (and other Bots) aempng to search for and reserve ckets
at the start of a sale, most Bots simultaneously aempt mulple connecons to the cket vendors’
systems, oen dozens or even hundreds at a me. These connecons are oen responsible for
sudden massive increases in Internet requests to ckeng websites when ckets are released.
To conceal that these large numbers of concurrent connecons originate from a single source,
most Bot users purchase hundreds or thousands of proxy IP addresses to use for each separate
connecon aempt. Moreover, because cket vendors block or slow trac connecons from IP
addresses that appear to be used by Bots, many Bots automacally rotate through their stores of
proxy IP addresses to bypass detecon and blocking. Bot users also typically register hundreds
or thousands of e-mail addresses, again to conceal that a single purchaser is responsible for
many concurrent transacons and to circumvent security measures designed to prevent Bots
from purchasing ckets.
With these capabilies, Bots can be extremely eecve for brokers. First, they can buy hundreds
of ckets in moments. NYAG has idened many instances in which Bots were able to purchase
hundreds of ckets within moments of the release of ckets to the general public or through a
pre-sale, including the examples shown in Figure 6.
Figure 6. Bots Buy Huge Numbers of Tickets Moments Aer Release.
Source: NYAG Invesgave Materials
1,012
tickets in
1 minute
U2 2015 Tour
Madison Square Garden
Bought by one bot on December 8, 2014, for a
July 19, 2015 concert.
520
tickets in
3 minutes
Beyoncé
Barclays Center
Bought by one Bot on March 4, 2013 for an
August 5, 2013 concert.
15,087 tickets in 1 day
U2 2015 Tour
Twenty Different Venues
Bought by two Bots on December 8, 2014 for
twenty concerts in the same tour across North
America.
522
tickets in
5 minutes
One Direction
Jones Beach
Bought by one Bot on April 14, 2012 for a June
28, 2013 concert.
19
Addionally, by being rst in line, Bots can also purchase the most desirable seats to many shows.
For example, NYAG reviewed a case study by one cket vendor showing that Bots purchased
more than 90% of the most desirable ckets to one show. As part of its own analysis, NYAG
found that, of the 251 ckets one Bot operaon purchased to a May 5, 2014 Coldplay concert
at the Beacon Theatre, 148 of those ckets were in the rst seven rows of the theater. This
amounted to more than 60% of seats in those seven rows.
As a result of these huge advantages, Bots are able to purchase many ckets to New York events.
The sources we interviewed uniformly stated that the usage of Bots has reached epidemic
proporons in the ckeng industry. Indeed, NYAG found that three brokers using Ticket Bots
collecvely purchased more than 140,000 ckets to events in New York over a three-year period
between 2012 and 2014.
Broker Profile: The One-Man Bot Broker
Revenue: $1.4 million from events in New York (2014)
Method: Bot mimicking Ticketmaster app & 1,000+ credit cards
Broker A was an illegal Bot user who represents a typical species of broker. The
company is a one-man operaon that opened when its proprietor was sll in
college. The company paid overseas programmers for sophiscated cket-buying
Bots designed to mimic the Ticketmaster iPhone app and defeat its an-Bot
defenses. To maximize its yield, and avoid cket limits, the company obtained more
than 1,000 credit cards, many of which used ficous names, each designed to
appear to Ticketmaster as a different individual.
Armed with its Bot and collecon of credit cards, the company directed legions of
purchase requests at popular shows; its acons are part of what led to the quick
sellouts. Its harvest was put up for resale on numerous plaorms, including Stub-
Hub and Vivid Seats, and its own site. While it made much of its money from New
York venues, the company is not licensed in New York.
20
21
b. Even without Bots, Brokers Use Industry Knowledge and
Relaonships to Gain an Edge Over Fans.
Brokers who do not use Bots nevertheless have several signicant advantages over the average
fan based on their knowledge of the industry and relaonships.
First, brokers in many instances take advantage of pre-sale events oered to holders of certain
credit cards. Brokers are aware of which credit cards can be used in pre-sales and open accounts
for these credit cards. Brokers also closely follow cket release schedules and are prepared to
purchase ckets at pre-sales, when the compeon for ckets is oen less than during a public
on-sale.
In many instances, arsts try to get ckets into the hands of fans by oering pre-sales to ocial
fan clubs, for example by issuing special codes to fan clubs that allow members to log into the
pre-sale on cket vendor sites ahead of the general on-sale. Unfortunately, brokers oen sign
up for these fan clubs, obtain the codes, and either use them themselves or sell them to other
brokers.
Second, Brokers have resources to aempt many concurrent connecons. Brokers understand
that it is important to try to reserve and purchase as many ckets as possible during the rst
moments of an on-sale. Many brokers therefore use mulple computers or mobile devices to
try to connect to ckeng websites at the same me. NYAG spoke with one broker who owned
100 mobile devices and claimed to use dozens of those at a me by himself to try to purchase
ckets. Some brokers employ dozens of people simply to have many humans trying to reserve
and purchase ckets at the same me.
Third, many Brokers maintain direct relaonships with venues and sports teams. In these cases,
brokers are able to purchase ckets without compeng for ckets through the ckeng website.
Barclays Center, for example, has provided several brokers who own season ckets to Brooklyn
Nets games with the opon to purchase several hundred ckets to most other events at the
venue, including all music concerts. More than 500 ckets to an Elton John concert at Barclays
Center were distributed to brokers in this manner.
22
c. Ticket Limits Are Not Regularly Enforced.
The consumer may be familiar with the cket limits that are encountered on the websites of
most cket vendors. Popular arsts and their promoters typically seek to have vendors, like
Ticketmaster or AXS, impose a limit on the number of ckets that any individual can purchase
for any event. For concerts, a typical limit is eight ckets per purchase; more popular concerts
set smaller limits. The theory behind limits is intuive and obvious: by liming the number of
ckets any one individual can purchase, ckets will be more fairly distributed. If cket limits
were truly eecve if it were impossible for any one person to buy more than the stated
number of ckets – broker acvity would be greatly limited.
Our invesgaon suggests that in the case of some cket vendors, including Ticketmaster, cket
limits do not always have the eect the public would expect. Ticketmaster implements cket
limits by restricng the number of ckets that can be purchased in a single transacon (a per-
transacon limit). It does not, however, restrict at the me of sale the number of ckets that can
be purchased by a user through mulple transacons (a per-person limit). In other words, when
an arst requests an eight-cket limit, Ticketmaster will permit a user to purchase only eight
ckets in a single transacon, but then will allow the same user to make addional purchases of
eight ckets each.
Broker Profile: Licensed Broker
Revenue: $25.4 million (2013)
Method: Employees buy ckets en masse
Broker C is a licensed �cket broker. In opera�on since the 1990s, the company has
become highly profitable under the new �ckets law. The company employs approxi-
mately 25 full-�me employees; each employee is paid a wage to keep his or her
own accounts and purchase, using his or her own American Express credit cards, as
many �ckets as possible for shows or events decided upon by management. The
company pays its annual licensing fees to the New York Secretary of State.
Broker C is an example of the 2007 law in opera�on. On the one hand, the
company pays its licensing fees, discloses its ac�vi�es, and is in apparent compli-
ance with most of the regula�ons. On the other hand, the company’s sizable annual
profits – $4.8 million in gross profit in 2013 – are direct evidence of the prices that
the public pays on the secondary �cket markets.
23
Ticketmaster can aempt to idenfy violaons of the cket limit aer ckets have been sold
by canceling those transacons that exceeded the limit the arst had requested. It will only
undertake this review, however, if an arst specically requests the audit. Yet many arsts seem
unaware of this fact. For example, a sophiscated representave of several top arsts playing the
largest arenas told NYAG they had been unaware that Ticketmaster required a separate auding
request to enforce limits the arst had already requested, and had therefore never made such
a request. Thus, although some arsts are trying to make ckets available to average fans by
pricing them below market value and seng cket limits, the lack of real-me enforcement per
user at the point of purchase (or thereaer) undermines their eorts.
The main beneciary of the unenforced cket limits is, of course, the broker industry. As repeat
players, the brokers know well that the limits are rarely enforced. Our interviews with brokers
and others in the industry, and our review of cket purchasing data, reveal that the limits,
enforced or not, are easy to evade. The eect of these shortcomings in enforcement of cket
limits is exacerbated where brokers use Bots to speed up their purchases above the limits, to
the detriment of ordinary members of the public. As a result, brokers have an advantage as
compared to ordinary buyers, who believe the phrase “cket limits” means what it says.
d. Brokers, With and Without Bots, Buy Up Many Tickets to New York Events.
Evidence that brokers are responsible for the purchase of large quanes of ckets can somemes
be seen in sales data that tracks the purchasing acvies of enes holding mulple credit cards.
For example, Figure 6 below shows American Express cardholders who purchased more than
$3 million in ckets through Ticketmaster for events in New York State and surrounding areas
using their American Express cards. A single broker used 149 dierent American Express cards
to make more than 38,000 purchases totaling over $12 million just from 2013-2015, while other
brokers racked up similarly shocking volumes of purchases, somemes using just a handful of
credit cards.
24
Figure 7. Twelve Brokers Each Purchased More than $3 Million in Tickets.
Source: American Express Purchases to Ticketmaster for Events
in New York & Surrounding Areas (2013-2015)
29
29. The methodology for this analysis is in the Appendix.
$12,251,282
throug h
38,366
transactions an d
149
cards .
$11,668,951
throug h
45,711
transactions an d
87
cards
.
$6,175,235
throug h
16,805
transactions an d
644
cards
.
$3,838,637
throug h
13,678
transactions an d
170
cards
.
$3,129,622
throug h
12,450
transactions an d
132
cards
.
$5,595,105
throug h
26,656
transactions an d
25
cards
.
$4,625,014
throug h
11,680
transactions an d
54
cards
.
$3,697,400
throug h
11,046
transactions an d
20
cards .
$3,592,798
throug h
10,039
transactions an d
23
cards .
$3,309,475
throug h
7,992
transactions an d
289
cards
.
$3,361,108
throug h
18,776
transactions an d
4
cards
.
$3,280,623
throug h
8,061
transactions an d
57
cards
.
25
e. Brokers Sell Tickets at Substanal Markups.
Brokers prot by selling ckets at a substanal markup over face value. NYAG studied six cket
brokers and found they marked up their ckets an esmated 49% on average, ranging by broker
from an average of 15% to 118%. Figure 8 depicts the markups charged by representave brokers
for their most protable shows.
Figure 8. Examples of Six Brokers’ Markups Show Large Prots.
Source: Transaconal sales data for select brokers (2011-2014)
30
30. The methodology for this analysis is in the Appendix.
$1,600
$815
$1,600
$350
$218
$218
$544
$443
$544
$250
$570
$693
$64
$101
$1,602
$79
$4,400
$3,650
$3,526
$1,249
$932
$925
$1,590
$1,155
$1,008
$1,306
$1,221
$999
$1,404
$1,351
$1,256
$7,244
$5,336
$4,635
$0 $1,000 $2,000 $3,000 $4,000 $5,000 $6,000 $7,000 $8,000
Barbra Streisand @ Barclays Center
NY Rangers v. LA Kings @ Madison Square Garden
Barbra Streisand @ Barclays Center
U2 @ Madison Square Garden
Stanley Cup Finals @ Madison Square Garden
Stanley Cup Finals @ Madison Square Garden
Eric Clapton @ Madison Square Garden
NHL Conf. Finals @ Madison Square Garden
Eric Clapton @ Madison Square Garden
Yusuf Islam/Cat Stevens @ Beacon Theatre
NHL Conf. Finals @ Madison Square Garden
NHL Conf. Finals @ Madison Square Garden
Ed Sheeran @ Madison Square Garden
Eagles @ Madison Square Garden
Pearl Jam @ Barclays Center
One Direction @ Madison Square Garden
Barbra Streisand @ Barclays Center
One Direction @ Beacon Theatre
Ticket Face Value Markup Amount
Custom Bot
Broker A
Custom Bot
Broker B
Bot Broker B
Non-Bot
Broker B
Non-Bot
Broker A
Bot Broker A
$94
$228
26
Figure 9. Brokers Charge Large Markups.
31
Source: Source: Transaconal sales data for select brokers (2011-2014)
Our analysis also reveals that the brokers that commanded the greatest markups in the cket resale
market were those that used Ticket Bots, and, moreover, that the most sophiscated, custom
Ticket Bots were associated with the highest markups. This may be because the sophiscated,
custom Bots were able to purchase the most desirable seats, which could be resold for the
highest prices.
f. Speculave Tickets Increase Confusion and Risk for Fans.
In early December 2015, ckets purporng to be for Bruce Springsteen’s 2016 “River Tour
began appearing on secondary sites like StubHub, TicketNetwork and Vivid Seats. Some ckets
were even adversed for inated prices of as much as $5,800 per cket. The catch? No such
ckets had yet been released. What brokers were selling were “speculave ckets,that is to
say, ckets that they planned, or hoped, to buy later.
When a speculave cket is sold, the broker then tries to purchase an actual cket, at a lower
price, to provide to the buyer. The broker pockets the dierence between the price at which he
pre-sold the cket and the price at which he later bought the cket. Speculave ckets harm
both consumers and the cket industry. In many cases, consumers are just defrauded they
do not receive the specic seats they paid for, instead receiving ckets for some other seats.
In some cases, consumers receive no ckets at all. Speculave cket sales also drive up prices
for consumers and oen cause widespread confusion and frustraon among consumers, who
wonder how ckets can appear on the resale market before ckets are even released to the
public.
In 2014, singer Eric Church expressed his frustraon with such ckeng: Theyre taking the
money, and theyre not even on sale. There’s no ckets on sale yet. They don’t really have them
they’re promising the fact that they can get them. Its a damn scam is all it is. Its the maa. In
the case of the 2016 Springsteen tour menoned above, NYAG asked StubHub, TicketNetwork,
and Vivid Seats to take down the speculave cket lisngs, and they all complied. This one-me
enforcement acon was eecve in that instance, but the problem is recurrent.
31. The methodology of this study appears in the Appendix.
Broker
Avg.
Cost
Avg.
Resale
Price
Markup
Highest
Markup
Amount Percent
Custom Bot A $145 $317 $172 118% 7,154%
Custom Bot B $104 $188 $84 81% 2,190%
Bot A $100 $137 $37 36% 637%
Bot B $105 $137 $32 30% 456%
Non-Bot Broker A $78 $101 $23 29% 1,537%
Non-Bot Broker B $111 $128 $17 15% 1,601%
27
g. The Market Structure Is Atypical.
Another nal point to keep in mind concerning this industry: the typical market structure is
skewed. For one, ckets are oen sold on the primary market for a face value that is below
market price. Although this may sound surprising, our invesgaon suggests it is common
and reects various factors. For one thing, maers of goodwill and reputaon in the sale of
ckets are important to certain performers. That is clearest, for example, for events where
the promoter has non-commercial goals – like Pope Francis, who distributed free ckets for his
public appearances – or at events like the New York City Centers annual Fall for Dance Fesval,
which includes world-renowned performers but sells ckets for just $15, “in keeping with the
Fesval’s commitment to make dance accessible to everyone.
32
Even more explicitly commercial acts somemes price their ckets below what they might charge
to make them accessible to younger fans or those of ordinary incomes. The band Fugazi, acve
over the 1990s and early 2000s, for example, long kept its cket prices at $5.
33
Pearl Jam, among
the most popular bands of the 1990s, set its cket prices at or below $20 to make it possible
for young fans to aend its concerts. As the band said in prepared tesmony before Congress,
Although, given our popularity, we could undoubtedly connue to sell out our concerts with
cket prices at a premium level, we have made a conscious decision that we do not want to put
the price of our concerts out of the reach of many of our fans.
34
Somemes a mixture of economic and non-economic moves is evident. A sports team, like the
Brooklyn Nets or New York Mets, might not want to be accused of gougingfans, and hoping
to build a loyal fan base, and so might price its ckets accordingly. Some sense of this came in
2009 when the New York Yankees experimented with much higher cket prices and were subject
to widespread cricism and unaering media coverage. The New York Post described the
Yankees’ new plan as greed-driven-and-delivered” and the prices as “insanely and obscenely
high.
35
Eventually, in the face of both cricism and poor sales, the Yankees lowered their prices
dramacally. Somemes even a prot-maximizing performer might set prices lower than market
value to guarantee a sellout and a full house for markeng purposes. Some of the promoters we
spoke with argued that a public sense of high demand and cket scarcity is necessary to create
strong demand for ckets. Hence, seng cket prices at a low level, so as to drive sales, may be
necessary to create the sense of a “sellout tour” that stokes demand to aend it.
Cheaper ckets could be very benecial to consumers, as long as they reap the benets. The
problem with this industry is that a middleman essenally takes the benets intended for the
consumer. In other words, the cket broker or reseller, able to buy up ckets before the public
gets a chance, prots from the lower priced cket – charging the consumer a much higher price
and pockeng the dierence. To further complicate the industry, the enes most able to stop
this pracce are either powerless or not economically incenvized to stop the pracce. Arsts
may want to avoid this from happening but have to rely on the venues to sell ckets. Venues
have an interest in selling out ckets and have lile incenve to put protecons in place. Ticket
32. Press Release, “New York City Center Announces 2015 Fall for Dance Festival” (July 29, 2015),
available at
https://www.nyci-
tycenter.org/content/misc/PressRelease-FFD15a.pdf.
33. John Pareles, “Review/Rock; Melodies Amid Rant, Thoughts Amid Rage, N.Y. Times (Mar. 7, 1991),
available at
http://www.
nytimes.com/1991/03/07/arts/review-rock-melodies-amid-rant-thoughts-amid-rage.html.
34. Hearing of U.S. House Comm. on Govt Ops., Subcomm. on Info., Justice, Transp., & Ag., 103d Cong. (June 30, 1994).
35. Phil Mushnick, “Pricing fans out of Stadium no good for Yankees or MLB, N.Y. Post (Apr. 27, 2014),
available at
http://nypost.
com/2014/04/27/pricing-fans-out-of-stadium-no-good-for-yankees-or-mlb.
28
vendors also have an interest in selling as many ckets as they can; they collect a fee with each
cket that is sold, a fee that is the same regardless of whether it is paid by you or a reseller.
B. High Fees for Unclear Purposes Raise Concerns
Next to sellouts, few issues seem to aggravate consumers as much as the addion of unexplained
convenience” charges or “service” fees to the prices of ckets. These fees can be of considerable
cost to consumers as a percentage of a ckets face value, or in absolute dollar terms. For
example, a cket to the Professional Bull Riding event at Madison Square Garden had $42 in fees
aached; a cket to
see Janet Jackson at
Jones Beach Theater
came with a $28 fee.
In 2010, the New
York legislature
amended the State’s
law governing cket
fees.
36
The statute,
as it stands, bans the
addion of
any
fee to a cket by a venue operator or its cket vendor except fees that are
associated with “special services” for which a “reasonable service charge” may be levied. The
statute states that special services include but are not limited to such services as “sales away from
the box oce, credit card sales
[37]
or delivery. Thus, charges added to a ckets face value violate
State law if they are either (1) mandatory, general fees, unconnected to the provision of “special
services,” or alternavely, when (2) such fees reach levels that are no longer “reasonable.
A ban on the charging of mandatory fees unconnected to a bona de service is a common consumer
protecon measure. New York City, for example, expressly bars restaurants from adding general
surcharges to the prices on menus.
38
Relatedly, the U.S. Department of Transportaon considers
it an “unfair and decepve pracce” to adverse cket prices that do not include mandatory fees
and taxes.
39
The New York law on cket charges creates a similar rule by banning the charging
of fees unrelated to the provision of specialized services, and also regulang the fees that are
charged.
36. N.Y. Arts & Cult. A. L. § 25.29.1. The Senate Sponsor Memorandum summarized the provision as follows: “requires that service
charges in association with tickets sold be reasonable.
37. Despite this provision, imposing a surcharge on credit card use is illegal under General Business Law § 518. (“No seller in any
sales transaction may impose a surcharge on a holder who elects to use a credit card in lieu of payment by cash, check, or similar
means”). See Expressions Hair Design v. Schneiderman, No. 13-4533 (2d Cir. 2015),
available at
http://law.justia.com/cases/fed-
eral/appellate-courts/ca2/13-4533/13-4533-2015-09-29.html.
38. New York City Rule § 5-59 provides: A seller serving food or beverages for consumption on the premises may not add sur-
charges to listed prices. For example, a restaurant may not state at the bottom of its menu that a 10 percent charge or a $1.00
charge will be added to all menu prices.
39. 14 CFR 399.84.
“I purchased a 30 dollar cket to see A View From
A Bridge. During the transacon I saw that a ten
dollar fee would be added for handling. Handling
what? I am using my own printer to print my own
cket!”
- Typical consumer complaint to NYAG about fees
1. Fees are Charged by Ticket Vendors and Venues.
Relying on publicly available informaon, NYAG conducted a broad study of the fees being
charged in New York State by 150 dierent venues and their cket vendors.
40
The examinaon
reveals substanal variaon in what fees are charged, but it also reveals some clear paerns.
First, it is worth nong that some outlets sell their ckets themselves; these venues are oen
smaller and non-prot. Examples include the Lancaster Opera House and the New York City
Ballet. These outlets or organizaons either do not charge any mandatory fees at all, or charge
a small service and/or handling fee for buying away from the cket oce that does not generally
vary with the price of the cket.
Most venues, however, sell their ckets pursuant to an exclusive contract with a cket vendor
like Ticketmaster or Tickets.com. This group includes the larger venues, such as Brooklyn’s
Barclays Center, Albanys Times Union Center, and Bualo’s First Niagara Center. However,
smaller venues, like the Ulster Performing Arts Center in Kingston and the Stanley Center for the
Arts in Uca, also use cket vendors. Generally speaking, the venue/cket vendor combinaons
charge fees and use a variety of formulas to set them, such as adding a xed charge based on
the type of cket or charging some percentage of the total price of the ckets.
While it is ckeng vendors (mainly Ticketmaster) that bear the brunt of the complaints about
fees, some venues play a central role in seng such fees and share in the money that the cket
vendors collect. Our review of long-term contracts between Ticketmaster and large venues like
Barclays Center or Madison Square Garden reveals that the pares negoate the fee amounts
including “convenience charges” or “service fees” (per cket) and “processing fees” (per order)
that can be shared between the venue and cket vendor.
41
The venue may also ban the cket
vendor from levying other charges, such as “delivery fees” for PDFs of ckets.
Our examinaon of the fees charged in New York also allowed us to esmate the average fee
amounts charged by the larger venues in combinaon with three cket vendors for events
in this State: Ticketmaster, TicketWeb, and Tickets.com. On average, combining the fees
charged by these three vendors and their venue clients for 150 New York venues, we found an
average surcharge of 21% of the face value of a cket, which amounts to almost $8 in fees on
average. However, our invesgaon also revealed that in certain cases, cket fees charged were
surprisingly high in real dollar terms, even if they were not a large percentage of the face value
of the cket. The following chart shows examples of ckets for several events in which the fees
exceeded $25 per cket.
40. The methodology of this study appears in the Appendix.
41. Ticket vendors often enter into multi-year contracts with venue operators to sell tickets to all events at the operator’s facility.
The terms of these contracts generally set forth the convenience charges and other fees that Ticketmaster must collect, and the
portions of those fees and charges that it must deliver to the venue, as well as the portions that Ticketmaster can keep.
29
30
Figure 10. Examples of Large Fees for Select Events and Venues.
Source: Fees collected from Ticketmaster, Tickets.com & TicketWeb (2015)
Professional Bull
Riding: Built Ford
Tough Series
Radio City
Christmas
Spectacular
Brooklyn Nets vs.
Phoenix Suns
Janet Jackson:
Unbreakable
World Tour
Black Sabbath:
The End
$42
in fees or 21%
of face value
$41
in fees or 11%
of face value
$30
in fees or 7%
of face value
$28
in fees or 16%
of face value
$28
in fees or
16%
of face value
31
2. Event Ticket Fees are Higher than Most Online Vendors.
Because Ticketmaster, TicketWeb, and Tickets.com all conduct a major poron of their business
online, where costs are oen lower than for brick-and-mortar retailers, we examined other online
plaorms that, similar to cket vendors, interact with consumers on one side and sellers on the
other side, to assess whether charging consumers large services fees separate and apart from
the ckets face value is nonetheless typical or perhaps necessary for ckeng. In this regard we
examined online vendors of ckets for airplane seats, such as Expedia and Priceline, and vendors
of ckets to lms, like Fandango and MovieTickets.com. We also examined online plaorms
that sell a broader mixture of goods, like Amazon and Etsy, to see if they charge consumer fees
separate and apart from the posted prices.
Our examinaon suggests that, among online plaorms, the vendors of event ckets appear to
charge fees to consumers that are higher than most other online vendors – in fact, most of the
online plaorms we examined charge no general” fees to consumers for the sale of ckets or
goods and post prices that already reect the full cost to the consumer. That is true of Amazon
and Etsy, and it is also typically true even of vendors of airline ckets, like Expedia and Priceline.
42
The main excepons are the sites that sell movie ckets, because those vendors charge consumer
fees that are close in magnitude to the average fees charged in live event ckeng. Fandango,
for example, charges moviegoers a “transacon fee” between 75 cents and $2.50 per cket, and
a “convenience fee” of $1.50 per cket.
It is unclear what services are covered by the “convenience charges,” “service fees, and
“processing fees” collected by online vendors of live event ckets, and why those services warrant
charging consumers fees that are higher than in most other online contexts. To the extent that
cket vendors and their venue operator clients collect these fees for anything other than the
provision of special services,they would be in violaon of New York law. In addion, any of
these fees that exceed a “reasonable” charge are impermissible under the law.
42. Rather than collecting money through fees to the consumers, these companies collect money from sellers who use their
platforms to sell to consumers. We observed no consumer fees being charged on any of these companies’ websites, and indeed
Expedia and Priceline state they rarely charge broker fees to consumers.
32
C. Restraints of Trade in Tickeng
Up to this point, the Report has focused on the sale and brokerage of ckets, which are subject to
industry-specic regulaon under Arcle 25 of New York’s Arts and Cultural Aairs Law. However,
under the Donnelly Act, the Aorney General is also tasked with enforcing the laws prohibing
unreasonable restraints of trade and the unlawful monopolizaon of markets. Pursuant to that
authority, NYAG has serious concerns about several ongoing pracces related to the sale and
resale of ckets in New York. In parcular, NYAG is concerned with the seng of price oors on
cket resale, the pracces that impede consumer access to alternave cket resale plaorms,
and, in parcular, the combined eect of such conduct on consumers. NYAG is involved in an
ongoing mul-state invesgaon of these issues.
1. Seng of Ticket Resale Price Floors.
Some cket issuers parcularly sports teams, including NFL teams and the New York Yankees
have put in place “price oors. Price oors are rules designed to prevent ckets from being
sold at a price below some level, usually the face value of the cket or something close to it.
For example, many NFL teams encourage or even require cket holders to use Ticketmasters
“NFL Ticket Exchange” plaorm which is frequently billed as the ocial resale site and the
only “safe” place to buy secondary NFL ckets. However, on NFL Ticket Exchange the seller is
then prohibited from seng a price below a certain level generally the face value of the cket.
Consequently, if a season cket holder wants to sell his or her ckets for a lower price than he
or she paid, he or she may be unable to do so, or may need to use a complex procedure to do so
on another plaorm.
NYAG’s concern with price oors is twofold. First, buyers of ckets on Ticket Exchange or other
sites with price oors are frequently not informed that the ckets they are buying are subject
to a oor. It is therefore easy for buyers to be fooled into believing what they are paying is the
market price for a cket, when in fact the buyer is paying a price arcially inated by a price
oor. The more aggressively sports leagues and individual teams push cket buyers and sellers
to use their “ocial” secondary markets, the more serious this problem becomes.
More fundamentally, even when buyers are informed of price oors, the oors deprive the
public of a chief benet of the market-driven approach taken by the 2007 law: lower prices. In
parcular, price oors may make it impossible to obtain ckets on the team-promoted Ticket
Exchange plaorm for below face value when demand decreases. As described above, when
contemplang the legalizaon of cket resale, the sponsors of the 2007 legislature repeatedly
expressed the hope that legalizing prot resale might lower prices for consumers. A clear source
of such savings is lower prices when demand falls. For example, near the end of an unsuccessful
baseball season, the ckets to watch a team not desned for the playos may go down sharply,
allowing fans who otherwise might not be able to aord to see a match to buy ckets for far less
money.
Overall, NYAG believes there is lile to say in favor of price oors: They tend to expose the public
to the full costs of the new cket economy, while depriving the public of the benets.
33
2. Pracces That Impede Consumer Access to Alternave or “Unocial”
Ticket Resale Plaorms.
The problems that come with price oors are exacerbated by eorts to push consumers to the
resale plaorms that maintain them and pracces that impede consumer use of plaorms that
do not. NYAG is therefore greatly concerned with other forms of restraints imposed on the resale
of ckets, parcularly when those restraints impede or complicate consumer access to resale
cket plaorms that do not enforce price oors. Examples of such pracces include delayed
delivery of PDF versions of resold ckets, and policies that place season cket holders at risk of
cancellaon of their cket subscripons when they sell on unocial resale plaorms.
The consumer harm caused by price oors is directly related to the success of the eorts to
pressure consumers to use only the primary sellers (usually Ticketmasters) plaorm. That is
because plaorms that use price oors are at a natural compeve disadvantage compared to
those that do not, and so if consumers engage in comparison shopping, they will gravitate to the
plaorms with a more open market. However, to the extent that the combinaon of pracces
such as PDF-delays and cancellaon threats pressure consumers into using the plaorms that
incorporate a price oor, consumers will be harmed. This is parcularly true given the inadequate
disclosure of the existence of these price oors. Without knowing that beer prices could be
available elsewhere, it is harder for consumers to benet from the compeve market.
34
RECOMMENDATIONS
The State regards ckeng and cket resale as a maer aected with the public interest.
43
Unfortunately, if one of the purposes of the 2007 law which repealed the ban on “scalping
– was to reduce cket prices, or make ckets more available to ordinary” fans, it has not been
successful. While there have been some posive developments, the basic reality described in the
1999 Report remains the same. If nine decades of experience have taught anything, it is that the
incredible demand for desirable ckets can be exploited even in the face of determined eorts
to get ckets in the hands of the public. The long history of cket scalping,the challenges
inherent in enforcement, and the incenves and prots available may make some think it is
impossible to x ckeng. Nonetheless, the situaon is not hopeless.
There are several concrete steps industry parcipants can take to make the system fairer and
more transparent. In light of the ndings in this Report, we believe that the legislature should
hold hearings on the subject of ckeng in New York to challenge key members of this industry
to put forth how they might help more ckets get into the hands of ordinaryfans. However,
if soluons are not implemented soon, the best and cheapest ckets will connue to go to the
resellers and consumers will connue to ask why they cannot get ckets.
A. Ticket Resale Plaorms Must Ensure Brokers
Comply With the Law.
Ticket resale plaorms such as StubHub, TicketsNow, TicketNetwork and Vivid Seats are the
major vehicles through which ckets are sold on the secondary market. They are thus in the best
posion to eect change in the industry. There are several steps these companies should take to
contribute to the success of this market.
First, each cket resale plaorm should ensure that professional resellers comply with New York
State’s licensing provisions by requiring that the resellers provide their cket reseller license
number as a condion of using the plaorm. In most cases, by examining the volume of resale
business a reseller conducts, resale plaorms can easily disnguish professional resellers – who
engage in the business of cket resale and are required to obtain a license – from fans that are
simply reselling ckets purchased for their own personal use – who are not required to have a
license.
Second, each cket resale plaorm should enable professional resellers to disclose the face
value of the ckets they resell, which is required of resellers by New York State law.
44
. It is
currently impossible for well-meaning resellers to disclose the face value of ckets on most resale
plaorms. Secondary market plaorms have the capability, however, of adding funconality to
their websites and apps to facilitate brokers’ compliance with the law, for example, by enabling
resellers to provide face value informaon when posng a cket for sale. They should do so.
43. N.Y. Arts & Cult. A. L.§ 25.01.
44. Id. § 25.23.
35
B. Industry Players Must Increase Transparency Regarding Ticket
Allocaons and Limits.
The precise manner in which ckets are distributed – to industry insiders through holds, to non-
public groups through pre-sales, and to the general public through public on-sales is rarely,
if ever, disclosed to the public. This informaon vacuum allows various players in the industry
to avoid their share of responsibility for the current problems the industry faces. It also breeds
mistrust from consumers, who feel like the system is xed against them.
To address these issues, industry players must publicly disclose the allocaon of ckets through
holds, pre-sale events, and public on-sales. This requires that many players, including promoters
and cket vendors, work together to eecvely distribute informaon to the public.
In addion, wherever cket vendors claim that cket limits are enforced, they should enforce
those limits as a maer of course on a per-person basis. If such limits are not actually being
enforced on a per-person basis, cket vendors must disclose that informaon.
C. Ticket Vendors Must Address the Bot Epidemic.
Although NYAG has taken steps to stop many of the largest operaons from using Bots to purchase
ckets to events in New York, we recognize that other brokers may take their place. A longer-
term soluon must include improvements in Bot detecon and prevenon methods. While
the industry works on long-term technological soluons, steps can be taken to reduce Bot use
in the near term. Towards that end, NYAG has analyzed the informaon and material gathered
through its invesgaon. Based on this analysis, we have compiled several recommendaons
for enhancing Bot detecon systems.
NYAG has contacted two of the largest ckeng companies, Ticketmaster and AXS, to discuss
our concerns with Bots and cket limits, and suggested concrete reforms to their operaons
that, we believe, will substanally reduce the acvies of illegal brokers and make the sale of
ckets fairer. Some of the suggesons included preempvely enforcing cket limits, analyzing
purchase data to idenfy ongoing Bot operaons, and invesgang resellers regularly oering
large numbers of ckets to popular shows, among others.
D. The Legislature Should Act.
To curb abuses in the industry and create a more equitable process for fans to obtain ckets, the
legislature should take the following acons:
i. Mandate the industry reforms outlined above.
In many cases, industry players do not have an incenve to reform. To ensure that the steps
described above are implemented in a meaningful and lasng way, the legislature should
mandate them.
36
ii. End the ban on non-transferrable paperless ckets.
“Paperless ckets” are non-transferrable event ckets that are designed to avoid resale by
brokers. Their dening feature is their non-transferrable nature, not the fact that they are
electronic, because many events oer ckets as PDFs or in other electronic form that the buyer
can sell or give to another person to use to aend the event. Non-transferrable paperless
ckets bear some resemblance to airline ckets, requiring the presentaon of idencaon
and the credit card used to buy the cket at the entrance to the venue. As such, it is harder
for paperless ckets to be transferred or resold, given the need to present the purchasing
credit card. Advocates of non-transferrable paperless ckets, including Ticketmaster and
many prominent performers such as Bruce Springsteen and Miley Cyrus, have argued that
they are an important and eecve “an-scalpingmeasure because increasing the diculty
of transferring ckets makes it harder for brokers to resell ckets for prot.
In 2010, New York amended its ckeng law to bar the issuance of non-transferrable
paperless ckets
unless
consumers are provided with an opt-out, that is, the opon of
another easily transferable ckeng format at no extra cost. The law is a
de facto
ban,
given that the opon allows brokers to request resalable ckets, and therefore erodes the
purported funcon of the non-transferrable paperless cket. As such, the change was widely
reported as a ban, and appears to be understood that way, at least by some performers. For
example, in September of 2014, popular musician Cat Stevens (now known as Yusuf Islam)
canceled a show in New York City on account of the ban. “I have been a longme supporter
of paperless ckets to my shows worldwide and avoiding scalpers,” he said. “Unfortunately
NY has a state law that requires all ckets sold for shows in NYC to be paper, enabling them
to be bought and sold at inated prices.
NYAG believes that New York’s ban on non-transferrable paperless ckets – the only one in
the naon – should be reconsidered and repealed. It is true that paperless ckets have met
with some resistance; they can somemes be inconvenient to consumers and venues (for
example, if the credit card is forgoen, or if someone wants to give a cket as a gi). That
may be why, naonwide, the use of such ckets has been limited to those acts that regularly
can expect to sell out. At the same me, paperless ckets appear to be one of the few
measures to have any clear eect in reducing the excessive prices charged on the secondary
markets and increasing the odds of fans buying ckets at face value. At least one major
arsts representave and one ckeng vendor reported seeing the use of non-transferrable
paperless ckets for the most desirable seats greatly reduce resale of ckets for those seats.
Allowing these types of ckets would therefore make it more dicult for brokers to connue
hoarding ckets and demanding exorbitant markups from fans. Moreover, such a change
would not require arsts, promoters or venues to use paperless ckets, but would merely
give them the opon to use them.
It is true that paperless ckeng is not a completely foolproof countermeasure. Brokers
wishing to resell such ckets can do so, but must then hire people to go to the concert and
physically walk fans through the gate at a show (assuming the ID provision is unenforced),
or alternavely, buy ckets with a prepaid gi” credit card that is physically mailed to the
consumer. Nonetheless, our conversaons with promoters of performers who use paperless
ckets and others lead us to believe that paperless ckets can eecvely limit the ability of
brokers to resell at great mulples on secondary markets. That may explain why, despite sll
represenng a small percentage of total cket sales, a number of high-prole bands now
use non-transferrable paperless ckeng outside of New York, including Bruce Springsteen,
37
Radiohead, Iron Maiden, Cat Stevens, Paramore, Widespread Panic and Metallica.
NYAG supports repeal of the
de facto
paperless cket ban, but a repeal raises several
concerns. Among the most important concerns raised against allowing non-transferrable
paperless ckets when the legislature passed the ban in 2010 was that they might serve a
dual purpose; they might not merely reduce broker markups, but also become a means by
which Ticketmaster might leverage its market power over primary ckeng for given events
into the dominaon of secondary sales as well. This concern follows because, in some
implementaons of paperless ckeng, ckets would not be completely non-transferrable.
Instead, ckets would be non-transferrable unless the cket was resold through Ticketmasters
own Ticket Exchange resale plaorm. When Ticketmaster is the exclusive” resale plaorm,
the potenal for extending its market power is obvious.
Nonetheless, the potenal for abuse does not jusfy a complete ban on paperless ckets.
When repealing the law, the legislature should consider safeguards that would permit non-
transferrable paperless ckeng while migang compeve concerns.
iii. Impose criminal penales for Bot use.
There is consensus in the cket industry that Ticket Bots have no place in a fair and equitable
cket market. Current law prohibits the usage of Bots and imposes civil sancons for
violaons. However, the prospect of criminal prosecuon may well have a greater deterrent
eect. Passage of the BOTS Act by Congress and the criminal penales that it would impose
for the use or sale of bots could also act as a deterrent.
iv. Cap permissible resale markups.
Unl 2007, ckeng laws capped the price of ckets sold on the secondary market. At rst,
for certain events, it was capped at two dollars above face value. Eventually, the two-dollar
limit became a 20% or 45% price cap (depending on the size of the venue). In 2007, the caps
were lied. Unfortunately, given the complexies of the market and the skewed incenves
that exist, under the current legal regime brokers will connue to use whatever means they
can, legal or illegal, to obtain ckets. New York should therefore reinstute a reasonable
limit on resale markups. Reinstang caps on markups would sll allow brokers a role in the
market but would also ensure that any price markups be reasonable.
38
APPENDIX – METHODOLOGY FOR NYAG ANALYSES
Holds & Pre-Sales Analysis
NYAG analyzed cket allocaon data for the highest grossing concerts in New York produced
by AEG, Live Naon, and Ticketmaster. NYAG’s analysis aggregated the data into four main
categories: holds, pre-sales, general on-sales, and other on-sales.
AEG
AEG produced cket allocaon data for the highest grossing concerts between January 1, 2012
and April 30, 2015 for which AEG served as the promoter at the following venues: Barclays
Center, Beacon Theatre, Madison Square Garden, Nassau Coliseum, Nikon at Jones Beach Theater,
Saratoga Performing Arts Center, The Theater at Madison Square Garden, Yankee Stadium, and
Icahn Stadium.
Data concerning the following concerts were excluded from the analysis due to observed data
inconsistencies:
• American Idol 2012 at Nassau Coliseum on August 22, 2012
• Carrie Underwood Fall 2012 at Nassau Coliseum on November 30, 2012
• Dane Cook at Beacon Theatre on September 14, 2013
• Enrique Iglesias at Madison Square Garden on September 25, 2014
• Enrique Iglesias Nassau Coliseum on September 14, 2014
Fresh Beat Band (Both Shows) at The Theater at Madison Square Garden on January 17,
2015
• Fresh Beat Band at The Theater at Madison Square Garden on November 29, 2013
• George Lopez at Beacon Theatre on April 25, 2014
• Jennifer Neles at Beacon Theatre on March 5, 2014
• Jusn Bieber at Barclays Center on August 2, 2013
• Kanye West at Madison Square Garden on November 23, 2013
• Kanye West at Madison Square Garden on November 24, 2013
• Mindless Behavior (Both Shows) at Beacon Theatre on July 13, 2013
• Nashville at Beacon Theatre on April 29, 2015
• Nashville at Beacon Theatre on April 30, 2015
• Nick Cave & The Bad Seeds at Beacon Theatre on March 28, 2012
• Nick Cave & The Bad Seeds at Beacon Theatre on March 29, 2012
• Nick Cave & The Bad Seeds at Beacon Theatre on March 30, 2012
• Pentatonix at The Theater at Madison Square Garden on March 18, 2015
• R. Kelly at The Theater at Madison Square Garden on November 21, 2012
• Sytycd at Beacon Theatre on October 30, 2014
• The Who Performing Quadrophenia at Madison Square Garden on November 14, 2012
• The Who Performing Quadrophenia at Barclays Center on December 5, 2012
• Trey Songz at The Theater at Madison Square Garden on December 5, 2012
39
For 42 of the remaining concerts, the cket allocaon data for each concert were aggregated in
the following way:
Total Tickets: The total number of ckets that were made available through a hold, pre-
sale, or general on-sale event. This gure was calculated by subtracng the Producon
Holds gure from the Starng Capacity gure. AEG denes Starng Capacity as: “…the total
number of ckets to be oered for sale as of the day before the rst pre-sale. Addional
ckets may be subsequently released for sale.AEG denes Producon Holds as: “… the
number of ckets held back from sale due to producon elements that create obstructed
views or eliminate physical seang. Tickets are oen periodically released before the show
typically for general public purchase from this total.
Total Holds: The total number of ckets placed on hold for an event. Calculated by
aggregang the following AEG categories: “arst holds, “agent, “venue, suite relos”
(“seats provided to venue’s suite holders due to obstrucon of suites”) AEG,“sponsor,
“label,” and “markeng.” Tickets in the Producon Holds group were not included.
Total Pre-Sale: The total number of ckets made available through pre-sales events.
Calculated by aggregang the following AEG categories: “Chase,Amex,“Ci,” “Fan Club,
“Fan Club / Gilt,” and “Venue Club.
Total Other On-Sale: The total number of ckets made available through the following
cket groups: “ADA,” “VIP Packages / Planum,” and “VIP Packages.
Total General On-Sale: The total number of ckets made available to the general public.
Calculated by subtracng Total Holds, Total Pre-Sale, and Total Other On-Sale from Total
Tickets.
Due to dierences in the producon of data, for two Katy Perry concerts at Barclays Center on
July 24, 2014 and July 25, 2014 the cket allocaon data for each concert were aggregated in
the following way:
Total Tickets: The total number of ckets that were made available through a hold, pre-
sale, or general on-sale event. Used the sum of Total Holds, Total Pre-Sale, Total Other On-
Sale, and Total General On-Sale, as dened below.
Total Holds: The total number of ckets placed on hold for an event. Calculated by
aggregang the following AEG categories: Arst and Supporng Acts,AEG,“CAA ‘MGMT,’”
“Label,” “Charity,” “Markeng Comps,” “Reviewer Comps,” and “Sponsor.
Total Pre-Sale: The total number of ckets made available through pre-sales events.
Calculated by aggregang the following AEG categories: “Cibank Pre-Sale,“Facebook,
AMX BluHold,” “Venue Club + Nets Basketball,VIP Pack 1-5,” “Planum,” and “Flex Hold.
AEG described Flex Hold as follows: “these are seats not yet priced. Once price is determined
they are released for sale during the pre-sale or general on sale.
Total Other On-Sale: The total number of ckets made available through the AEG category
“220° Sell” and “Flex Hold 2”. AEG describes 220° Sell as follows: “these are seats past the
180 degree full view sell line. Typically sold as side view,’ they are opened once the full view
inventory is depleted.
40
Total General On-Sale: The total number of ckets made available to the general public.
Uses the AEG category: “Esmate Of Total Opens Available For General On Sale,which is
described as follows: “This Is Inventory Held Back From The Pre-Sale To Be Sold Only at the
general on sale.
Live Naon/Ticketmaster
Live Naon and Ticketmaster produced cket allocaon data for the highest grossing concerts
between January 1, 2012 and December 31, 2013 for which Live Naon served as the promoter
at the following venues: Barclays Center, Beacon Theatre, Madison Square Garden, Nassau
Coliseum, Nikon at Jones Beach Theater, Saratoga Performing Arts Center, the Theater at Madison
Square Garden, Yankee Stadium, and Icahn Stadium. NYAG analyzed the data that was produced
as described below.
Live Naon also produced cket allocaon data for the highest grossing concerts between
December 10, 2013 and August 31, 2015 for which Live Naon served as the promoter for the
following venues: Barclays Center, Beacon Theatre, Madison Square Garden, Nassau Coliseum,
Nikon at Jones Beach Theater, Saratoga Performing Arts Center, the Theater at Madison Square
Garden, Yankee Stadium, and Icahn Stadium. The data produced by Live Naon, however, did
not contain sucient informaon to use in the NYAG’s analysis.
Data concerning the following concerts were excluded from the analysis due to observed data
inconsistencies:
• Eagles at Madison Square Garden on November 8, 2013
• Eagles at Madison Square Garden on November 9, 2013
• Eagles at Madison Square Garden on November 11, 2013
• Macklemore at Madison Square Garden Theatre on November 13, 2013
• Macklemore at Madison Square Garden Theatre on November 14, 2013
• Macklemore at Madison Square Garden Theatre on November 15, 2013
For the remaining 32 concerts, the cket allocaon data for each concert were aggreggated in
the following way:s
Total Tickets: The total number of ckets that were made available through a hold, pre-sale
or general on-sale event. Calculated by subtracng all cket groups idened as a producon
hold, such as “Rearview, “Producon Kills, and “Producon Holds, from the Total Tickets
Available” gure.
Total Holds: The total number of ckets placed on hold for an event. Calculated by aggregang
the following groups and others: Arst(s), “Management, “Support Act, Venue, Tour
Sponsor,“Label,“Live Naon,“Sponsor,“Markeng,“Promoon,Tour Hold,Arst Agent,
and “Live Naon Touring. Ticket groups idened as a producon hold, such as “Rearview,
“Producon Kills,” and “Producon Holds,” were not included.
Total Pre-Sale: The total number of ckets made available through pre-sales events. Calculated
by aggregang the following groups and others: “Fan Club OnSale, “Fan Club Presale, “LN/
Facebook, Venue Presale, Arst Presale, VIP1-2 Presale, “LN & Venue Presale, “Live
Naon Presale,“Season Tickets,“Chase Presale,“Mastercard OnSale,“Ci Preferred,AMEX
PreSale,” “AMEX Postsale,” and “CiCard Postsale.
41
Total Other On-Sale: The total number of ckets made available through the following cket
groups and others: “VIP Packages,” “Planum,” and “ADA Seang.
• General On-Sale: The total number of ckets made available to the general public. Calculated
by aggregang the following groups: “Public Onsale,“Public Protect Opens,“SRO-Opens Sell
Last,” “Dress,” and “Café Sell Last.
Analysis of Brokers Using American Express Data
American Express produced informaon on all transacons through a Ticketmaster merchant
account for New York State and surrounding areas for the period of April 1, 2013 to March 1,
2015. NYAG aggregated the individual transacons for each primary account holder to calculate,
for each account holder, the total number of transacons, the total cost of these transacons,
and the total number of unique cards used. NYAG then idened those account holders that
had completed purchases totaling more than $3 million.
Broker Markup Analysis
NYAG analyzed transaconal sales data for events in New York produced by six cket brokers to
esmate the margins they typically obtained in connecon with their resale of ckets. The data
cover sales by these six brokers made from 2010 to 2014 and include over 90,000 transacons.
All reported transacons for which a sales price and cost per cket that exceeded one dollar
were idened in the data were included in the analysis. The analysis esmated a markup for
each transacon, which is dened as the rao between the net revenue earned by the sale of
an individual cket and its cost. The average broker markups were then calculated across all
transacons and weighted by each transacon’s face value.
Fees Analysis
NYAG conducted a study of cket fees in an eort to esmate the typical fee amount charged
by cket vendors for entertainment ckets. The study reected fees typically charged by these
cket vendors as of October/November 2015. The study did not include every cket sold by cket
vendors in the State, but rather relied on informaon displayed on Ticketmaster, Tickets.com,
and TicketWeb’s websites in connecon with the sale of ckets for events in popular venues. In
parcular, the study collected fee informaon and cket prices for 150 New York venues listed by
the three cket vendors on their websites. For each venue, fee informaon was collected for up
to three randomly chosen events and, for each event, informaon was collected for every cket
category (e.g., orchestra, mezzanine). In total, a sample consisng of 859 dierent ckets was
created for venues explicitly listed or promoted by cket sellers. Fee informaon included any
costs associated with an online sale and delivery of a cket separate from its face value. These
costs may include service fees, credit card fees, facility fees, and/or any other fees associated
with the online sale and delivery of an event cket.
42
43